INDEX

Explanations

words related to illness and medical conditions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WEBPACK

-0.82

 Bourgoin

-0.76

 Humphries

-0.74

äinen

-0.72

 Réponses

-0.72

atop

-0.70

emia

-0.69

andExpect

-0.68

>>>>>>>

-0.67

phosa

-0.66

POSITIVE LOGITS

 Sill

1.06

llllll

1.06

ll

1.01

 Kall

1.00

llll

0.98

lll

0.98

lllll

0.97

Oll

0.96

LLL

0.96

llllllll

0.92

Activations Density 0.569%

words related to illness and medical conditions

No Comments

No Known Activations